基于聚类的空间匿名化位置隐私保护研究

In spatial crowdsourcing, tasks need to be completed by workers near the task location. In order to achieve efficient task allocation, workers usually need to upload their location information to crowdsourcing server. Only when workers\' location privacy is protected, can more workers be attracted to participate in task allocation. In this paper, we propose a privacy preserving framework to protect workers and the task location information in the task allocation stage. The framework transforms the workers and task location without affecting the distance calculation. At the same time, it anonymizes the workers\' location based on spatial anonymity technology. In the process of anonymization, we generate a certain amount of noise data according to the distribution of workers to further protect workers location privacy. Experiments on real-word datasets verify the effectiveness of the framework.